The orexin story and orexin receptor antagonists for the treatment of insomnia

Insomnia is present in up to one third of the adult population worldwide, and it can present independently or with other medical conditions such as mental, metabolic, or cardiovascular diseases, which highlights the importance of treating this multifaceted disorder. Insomnia is associated with an abnormal state of hyperarousal (increased somatic, cognitive, and cortical activation) and orexin has been identified as a key promotor of arousal and vigilance. The current standards of care for the treatment of insomnia recommend non‐pharmacological interventions (cognitive behavioural therapy) as first‐line treatment and, if behavioural interventions are not effective or available, pharmacotherapy. In contrast to most sleep medications used for decades (benzodiazepines and ‘Z‐drugs’), the new orexin receptor antagonists do not modulate the activity of γ‐aminobutyric acid receptors, the main inhibitory mechanism of the central nervous system. Instead, they temporarily block the orexin pathway, causing a different pattern of effects, e.g., less morning or next‐day effects, motor dyscoordination, and cognitive impairment. The pharmacokinetic/pharmacodynamic properties of these drugs are the basis of the different characteristics explained in the package inserts, including the recommended starting dose. Orexin receptor antagonists seem to be devoid of any dependence and tolerance‐inducing effects, rendering them a viable option for longer‐term treatment. Safety studies did not show exacerbation of existing respiratory problems, but more real‐world safety and pharmacovigilance experience is needed. This review provides an overview of the orexin history, the mechanism of action, the relation to insomnia, and key features of available drugs mediating orexin signalling.

molecule dual orexin receptor antagonists (DORAs) for the treatment of insomnia has led to the approval of suvorexant, lemborexant, and daridorexant (FDA, 2014(FDA, , 2019(FDA, , 2022)).These drugs share similar pharmacological properties in animals and humans, that is, they penetrate the blood-brain barrier and function as competitive orexin receptor 1 and 2 (OX1R and OX2R) antagonists with similar potency on both receptors.
DORAs block the binding of the wake-promoting neuropeptides orexin-A and -B to OX1R and OX2R and reduce wake drive.
1.1 | History and background: discovery of the orexin system The discovery of the orexin neuropeptides and their receptors dates back to 1998, when two independent groups reported their findings on this new signalling pathway.Researchers from the University of Texas Southwestern, led by Yanagisawa (Sakurai et al., 1998), named these peptides orexin-A and -B because they were originally thought to promote feeding (Ορέξη [orexis], Greek for appetite), while a group of scientists from Scripps Research Institute in La Jolla, led by Sutcliffe (De Lecea et al., 1998) named the peptides hypocretin-1 and hypocretin-2 because they are produced in the lateral/paraventricular hypothalamus and have some similarities to the gut hormone secretin.
In 1999, after the deorphanization of orexin receptors, the link between a dysfunction of the orexin system and narcolepsy was revealed with a first publication showing that a defect in the gene coding for the OX2R was responsible for canine narcolepsy (Lin et al., 1999) and a second demonstrating that orexin knockout mice exhibit a narcolepsy phenotype markedly resembling the human disease (Chemelli et al., 1999).In 2000, the confirmation was published that narcolepsy with cataplexy (sudden loss of muscle tone) in humans is the result of a deficiency of the orexin system characterised by the loss of orexin neurones (Peyron et al., 2000;Thannickal et al., 2000).
Narcolepsy with cataplexy is a sleep disorder characterised by severe daytime sleepiness and abnormal rapid-eye-movement (REM) sleep manifestations (Dauvilliers et al., 2007) and is described as a disorder of wake/sleep instability (Sorensen et al., 2013).
Over the years, a wealth of literature demonstrated that, indeed, the orexin system is a key player in wakefulness stabilisation (Saper et al., 2005).
These new insights into the biology of sleep-wake regulation sparked a dynamic interest of the pharmaceutical industry to develop drugs inhibiting the orexin system.It was hypothesised that the hyperarousal model (i.e., an imbalance/overactivity of the sleep-wake regulation) is the common pathway of the pathophysiology of chronic insomnia as defined by the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) criteria (Levenson et al., 2015;Riemann et al., 2010).
In 2007, dose-dependent sleep-promoting effects not only in rats and dogs, but also in humans were published for the DORA almorexant (Brisbare-Roch et al., 2007), which was followed by the first demonstration of clinical proof-of-concept (Hoever et al., 2012), that is, sleep promotion in patients with insomnia.Although development of almorexant was discontinued (Hoch et al., 2013), these results significantly accelerated the interest of Research and Development (R&D) organisations in this field, leading to several compounds subsequently entering clinical development, culminating in suvorexant being the first DORA receiving United States Food and Drug Administration (FDA) approval in 2014 (Jacobson et al., 2014;Michelson et al., 2014).
Interest in the development of orexinergic drugs remained high (or even increased), as a wealth of preclinical literature (Ramirez et al., 2013;Tannenbaum et al., 2014Tannenbaum et al., , 2016) ) indicated that DORAs promote sleep with fewer side-effects than classical hypnotics and preserve the natural sleep architecture.Classical hypnotics (e.g., benzodiazepines and 'Z-drugs') enhance the activity of the γaminobutyric acid type A (GABA A ) receptor, thereby broadly inhibiting the central nervous system (CNS) and numerous brain functions, resulting in, e.g., decreased motor function, impaired memory and cognition, potential for abuse and dependence, and, specifically for longer-acting benzodiazepines, zopiclone, and higher doses of zolpidem, next-morning hangover effects (Buscemi et al., 2007;Zammit, 2009).Bizarre behavioural effects including sleepwalking, sleepdriving, and making telephone calls have prompted warnings on the prescription, dispensing, and use of Z-drugs (Gunja, 2013a(Gunja, , 2013b)).

| The orexin system: a key regulator of wakefulness
The neuropeptides orexin-A and orexin-B are derived from the same precursor peptide (pre-pro orexin) (Sakurai et al., 1998).These peptides are exclusively produced in the hypothalamus by a discrete number of neurones ($50-80 Â 10 3 cells), and, although representing a restricted group of cells, they project widely into many brain regions (Peyron et al., 1998).The orexin peptides bind to the two related Gprotein-coupled receptors, OX1R and OX2R expressed in the targeted brain regions.Orexin-A binds to both receptors with similar affinity, whereas orexin-B binds the OX2R with higher affinity than OX1R (Sakurai et al., 1998).The orexin system is highly conserved across species (Sakurai et al., 1998).
The orexin system is recognised as crucial for the stability of wakefulness and the regulation of vigilance in the context of various physiological processes (De Lecea, 2012;Sakurai, 2007;Scammell et al., 2017).By integrating information from the internal and external environment, such as emotional state, light/dark cycle, sleep pressure, and energy balance, the orexin system adapts the arousal level to allow an appropriate behavioural response (Inutsuka & Yamanaka, 2013).Orexin-A levels in the brain extracellular and cerebrospinal fluid follow a circadian rhythm, rising during the wake period and dropping rapidly during sleep.Increasing orexin levels are necessary to compete with the increasing sleep pressure that builds up during long periods of wakefulness and to avoid falling asleep (Gotter et al., 2013;Modirrousta et al., 2005;Zeitzer et al., 2003).However, the pattern of activity of the orexin neurones does not only follow a circadian/arousal rhythm but also depends on the emotional state, with high activity recorded during positive emotion or social interaction (Blouin et al., 2013).
To stabilise wakefulness, orexin neurones send excitatory projection to wake-promoting neuronal populations (Hagan et al., 1999;Lee et al., 2005;Scammell et al., 2017).Figure 1 schematically depicts wake-promoting areas that inhibit and are inhibited by sleep-active centres such as the ventrolateral preoptic area (VLPO).This mutual inhibition could lead to abrupt transition between sleep and wakefulness without the reinforcing effect of the orexin system on wakepromoting areas, which prevents falling asleep at any time (Sakurai, 2007;Saper et al., 2005).Recently, De Luca et al. (2022) demonstrated that the orexin system also promotes arousal by inhibiting sleep.Indeed, orexin neurones stimulate GABAergic interneurones in the VLPO leading to inhibition of the so called 'VLPO sleep centre'.
Before DORAs were approved, there has been some debate based on preclinical work, whether dual or selectively blocking OX2R signalling is preferred to treat sleep-related disorders (Dugovic et al., 2009;Mang et al., 2012;Morairty et al., 2012).Indeed, early clinical trials using the selective OX2R antagonist seltorexant (JNJ-42847922) have shown that inhibition of the OX2R alone is sufficient to improve sleep variables (falling and staying asleep) in humans (Brooks et al., 2019;De Boer et al., 2018).Robust evidence from large trials in patients with insomnia is still missing, and, therefore, no firm conclusions can be drawn on whether the concept of selectively blocking OX2R will be a relevant treatment option for insomnia.

| INSOMNIA AND OREXIN SIGNALLING
Insomnia is described as a disorder of hyperarousal processes, including hyperactivation of the stress system (Levenson et al., 2015;Palagini et al., 2022;Riemann et al., 2010) and prevalence that increases with age (Morphy et al., 2007).Functional neuroimaging studies in patients with insomnia suggest that their wake-promoting areas may not be sufficiently inactivated during sleep, which would lead to difficulties initiating or maintaining sleep (Buysse et al., 2011;Nofzinger et al., 2006).The exact mechanisms leading to an imbalance between sleep-inducing and arousing brain activities are not fully elucidated, but a hyperactive orexin system might contribute to it (Riemann et al., 2010).The primary role of the orexin system is to adapt the arousal level to the behavioural need, as observed with the increase in vigilance in response to stress and emotions (Palagini et al., 2023;Soya & Sakurai, 2020).However, studies have shown that the orexin system could potentially modulate stress (Kaplan et al., 2022) and that DORAs were able to reduce fear-related behaviour in rodents (Steiner et al., 2012).Although there is not much direct evidence for a hyperactive orexin system in insomnia, its role on arousal and stress modulation supports the hypothesis that a dysregulation of the orexin system could contribute to the hyperarousal processes of insomnia.
Interestingly, in mice, it was shown that the number of orexin neurones decreases with age, but that the remaining ones become hyperexcitable, leading to a general decrease of the arousal threshold during sleep and sleep fragmentation (Li et al., 2022).Although it was not a model of insomnia, these data could shed light on the underlying mechanism leading to some of the sleep complaints experienced by the elderly.
Extensive preclinical literature exists on the potential role of the orexin system outside of sleep-wake regulation including a role in energy homeostasis, learning and memory, stress/emotion, reward and pain (Latifi et al., 2018;Razavi & Hosseinzadeh, 2017;Villano et al., 2022).
However, the translation from animal to human and the therapeutic potential of targeting the orexin system thus far has only been demonstrated in the regulation of sleep and wakefulness.Indeed, ORAs promote sleep in animals and today, three DORAs are approved treatments for insomnia in humans (Beuckmann et al., 2017;Cox et al., 2010;Roch et al., 2021), while the orexin receptor agonist danavorexton was shown to promote wakefulness and decrease narcolepsy symptoms both in animal models and in patients with narcolepsy (Evans et al., 2022).
Considering the growing evidence of insomnia as a disease of hyperarousal caused by an overacting wake system, the therapeutic approach of pharmacologically enhancing sleep-signalling (i.e., GABA A receptor modulation) might not be physiologically appropriate.
Instead, the recently emerging drugs of the DORA class, reversibly decreasing the excessive wake signalling, offer a viable alternative to pharmacologically treat insomnia disorder.

| OREXIN RECEPTOR ANTAGONISTS IN CLINICAL PRACTICE
Current treatment guidelines recommend cognitive behavioural therapy for insomnia (CBT-I) as the first-line treatment of insomnia (Qaseem et al., 2016;Riemann et al., 2017).A current CBT-I regimen comprises elements of stimulus control, sleep restriction, sleep hygiene, and cognitive therapy.However, in practice this treatment is not always available or desired by patients, and not all benefit from it.Adherence to CBT-I activities relies on intact self-discipline and motivation, which is often not possible or simply too demanding for patients.
Therefore, sleep medications are often prescribed (Schutte-Rodin et al., 2008).Nevertheless, there is currently no consensus with regard to which pharmacological intervention is most efficacious for a certain type of insomnia or which drug has the best benefit-risk profile (Perlis et al., 2022).Besides GABA A receptor modulators, ramelteon is a melatonin receptor agonist that has only moderate effects in shortening sleep latency and increasing subjective total sleep time (sTST), while it has little to no effect on wake after sleep onset (WASO), questioning its usefulness to treat sleep maintenance problems (Kuriyama et al., 2014;Simpson & Curran, 2008).
The emergence of a new and targeted class of sleep medication with the DORAs is opening a new avenue for the pharmacological treatment of insomnia.The three different DORAs share many characteristics and mainly differ in pharmacokinetic/pharmacodynamic (PK/PD) properties and the way their effects were investigated during clinical development in healthy subjects and in patients with insomnia (Xue et al., 2022).Although establishment of concentration-response relationships is notoriously challenging for sleep-promoting drugs, suitable PK characteristics are key, that is, fast absorption is pivotal for sleep onset, and quick disposition and reasonably fast elimination should provide beneficial sleep effects for a major part of the night without leading to residual drug effects the next morning (Mignot et al., 2022;Moline, Thein, Bsharat et al., 2021).This latter aspect receives increasing attention as patients desire to lead a normal life despite treatment with a sleep medication (participation in traffic, business, and social life).In this context, a third aspect of insomnia disorder besides sleep onset and/or sleep maintenance issues needs to be addressed, that is, impairment of daytime functioning, which includes fatigue, sleepiness, comorbid mood disorders, and cognitive impairment (Hudgens et al., 2021;Kyle et al., 2010).If prescription of a sleep-promoting drug can demonstrably lead to improved daytime functioning, this would be a great asset as it addresses one of the cardinal aspects of insomnia, which does not only affect patients' nighttime sleep but also their day (APA, 2013).

| Pharmacology, PK, and PD
Besides the similar in vitro potency on both orexin receptors, the marketed DORAs also share similarities in the metabolism and excretion in humans, that is, all three compounds are lipophilic drugs that are hepatically metabolised by cytochrome P450 3A4 (CYP3A4), and mainly excreted into faeces.In addition, the similar CYP3A4 substrate properties of the three DORAs is worth a mention.
In published drug-drug interaction (DDI) studies with the three DORAs (Boof et al., 2019;Landry et al., 2021;Wrishko et al., 2019), a more than two-fold (but less than five-fold) increase in total exposure upon co-administration with a moderate CYP3A4 inhibitor (i.e., diltiazem for suvorexant and daridorexant, fluconazole for lemborexant) was observed, which unequivocally qualified the DORAs as sensitive CYP3A4 substrates (Preskorn, 2023).Although there are no reports of increased intensity or prolonged duration of next-morning somnolence in patients or study participants upon co-administration of CYP3A4 inhibitors with a DORA, concomitant use of daridorexant and strong inhibitors of CYP3A4 is contraindicated in Europe (not recommended in the USA), while caution and dose reduction is advised when using a DORA in combination with a moderate CYP3A4 inhibitor (FDA, 2014(FDA, , 2019(FDA, , 2022)).In this context, it is important to note that not only drugs can cause interactions, but also certain nutrients have the potential of inhibiting (grapefruit or grapefruit juice) or inducing (St John's wort) CYP3A4 to a clinically relevant extent.Thus, the DDI propensity regarding concomitant consumption of grapefruit products is mentioned in the prescribing information of the DORAs (FDA, 2014(FDA, , 2019(FDA, , 2022)).
Figure 2 shows some differences in PK, in that daridorexant and suvorexant have a short to intermediate terminal elimination half-life (t ½ ) of 8 and 12 h, respectively, while the reported effective t ½ of lemborexant is 17 h (terminal t ½ is 55 h), resulting in a plasma accumulation ratio of 2-3 (Muehlan et al., 2020;Murphy, Yasuda et al., 2017;Preskorn, 2022).In theory, prolonged drug elimination or increased accumulation will lead to increased plasma levels that may carry the risk of daytime somnolence; however, an increased incidence of relevant hangover symptoms associated with the DORA class has not been reported (Rosenberg et al., 2021).
Effects on PD variables of different psychometric tests, evaluating vigilance and attention, motor coordination, and self-reported sleepiness or alertness were consistently observed in early clinical trials with the three DORAs following morning administration to healthy participants (Landry et al., 2020;Muehlan et al., 2018;Sun et al., 2013).The objective of these early trials was to evaluate the PK/PD properties and the tolerability and safety profile up to high supratherapeutic doses (e.g., for lemborexant, 200 mg was administered, a dose 20-fold greater than the highest dose [10 mg] approved for the treatment of insomnia) to define a safe and efficacious dose range to be tested in subsequent studies in patients with insomnia.
Considering the tolerability and safety aspects of these early trials in healthy subjects following daytime administration, the adverse event (AE) profile (e.g., headache and somnolence) may not directly apply to patients with insomnia, who will take the drug in the evening and benefit from its efficacy both on sleep and daytime functioning.

| Efficacy in patients with insomnia
With regard to the classification of insomnia, the different guidelines in use, i.e., DSM-5 (APA, 2013), the International Classification of Sleep Disorders (Sateia, 2014), the American Academy of Sleep Medicine Clinical Practice Guideline (Sateia et al., 2017), and the European guideline for the diagnosis and treatment of insomnia (Riemann et al., 2017), all consider insomnia as a chronic condition with the symptoms present for ≥ 3 months.In addition, insomnia is a clinically persistent condition.Shortcomings of the traditional hypnotics are that their use is limited to short-term treatment due to known safety concerns such as dependence and sometimes severe withdrawal symptoms (Morin et al., 2009(Morin et al., , 2020)).In contrast, all three approved DORAs have demonstrated sustained efficacy up to 12 months of treatment and are devoid of an increased safety risk, in particular no evidence of physical dependence, tolerance, or rebound has been reported upon treatment discontinuation (Kunz et al., 2022;Michelson et al., 2014;Yardley et al., 2019).
Suvorexant was the first representative of the DORA class, available for patients since 2014.It is important to mention that the two Phase III trials conducted in > 2000 patients (treatment duration 3 months) focused on the unapproved doses of 40/30 mg (nonelderly/elderly), while fewer patients were randomised to 20/15 mg (Herring et al., 2016), the latter doses eventually receiving marketing approval by the FDA.Results of the approved doses of suvorexant on the objective sleep variables latency to persistent sleep and WASO as measured by polysomnography, as well as the subjectively (i.e., based on patients' own perception) scored variables subjective latency to sleep onset and subjective WASO (sWASO) are presented in Tables 1   and 2. Similarly, results of the registration trials of lemborexant and daridorexant are presented in Tables 3-6.
An important, yet complex aspect in chronic insomnia is the dynamic nature of the symptoms such as difficulty initiating and/or maintaining sleep and/or early morning awakening, all fluctuating over time (Rosenberg et al., 2021) (Herring et al., 2016;Kärppä et al., 2020;Mignot et al., 2022;Rosenberg et al., 2019).The confirmatory studies with suvorexant and daridorexant have replicated these benefits in two similar independent studies of 3 months duration, although in the USA the efficacy profile of suvorexant is compromised by a starting dose of 10 mg, that was not investigated in Phase III, showing marginal efficacy on sleep onset insomnia (FDA, 2014).
Per definition, insomnia disorder is multifaceted and does not rely solely on night-time symptoms (APA, 2013;Rosenberg et al., 2019).
The assessment of an improvement in daytime impairment, which is a major symptom of insomnia, has been neglected in the development programme of former sleep medications.While improvement in sleep T A B L E 1 Efficacy of suvorexant 15/20 mg at Month 3: effects on objective sleep variables in patients with insomnia disorder.1 and 2), 291 non-elderly adults were treated with 20 mg and 449 with placebo, while 202 elderly patients were treated with 15 mg and 318 with placebo.Abbreviations: CI, confidence interval; LSM, least squares means; SD, standard deviation; sLSO, subjective latency to sleep onset; sTST, subjective total sleep time.These results suggest that the combination of a targeted mechanism of action together with a PK profile that covers the night as shown for the DORAs (Muehlan et al., 2020), can result in a clinical benefit not only on night-time but also on daytime symptoms of insomnia.

| Safety and tolerability
The approved DORAs share similarities in their safety profile.They are contraindicated in patients with narcolepsy, and, due to their interaction potential, in situations in which concomitant drugs that are strong inhibitors of CYP3A4 are required (e.g., the anti-infectives itraconazole or clarithromycin).
Narcolepsy type 1 is caused by severe loss of the wakestabilising orexin neurones (Nishino et al., 2000), which results in excessive daytime sleepiness and episodes of cataplexy (short periods of muscle weakness, typically elicited by an emotional trigger, e.g., laughing).Narcolepsy type 2 is of unclear aetiology but may result from a less severe loss of orexin neurones (Thannickal et al., 2009) and presents with similar symptoms as narcolepsy type 1 but without cataplexy (Scammell, 2015).Prolonged pharmacological blockade of orexin receptors by DORAs might aggravate LPS, min or LSGM ratio 5 mg 44.9 (36.5)À19.5 (33.1) 0.77 (0.67 to 0.89) a 10 mg 44.6 (33.0)À21.5 (32.4) 0.72 (0.63 to 0.83) a Placebo 43.9 (33.6)À7.9 (32.0)SEF, % 5 mg 68.4 (11.3) 12.9 (9.7) 7.1 (5.6 to 8.5) a 10 mg 67.9 (10.8) 14.1 (10.5) 8.0 (6.6 to 9.5) a Placebo 68.9 (9.6) 5.4 (9.9) narcoleptic symptoms in patients with already impaired orexin signalling.In rodents, under rewarding stimulation during the active phase, DORAs induced cataplexy-like events (Mahoney et al., 2020).Although cases of sleep paralysis and muscle weakness were observed in clinical trials, no evidence of an increased incidence of cataplexy was reported in patients treated with a DORA (Michelson et al., 2014;Mignot et al., 2022;Rosenberg et al., 2019).As per prescribing information of the marketed DORAs, safety warnings and precautions include CNS-depressant effects and daytime impairment including reduced motor coordination that could affect executive functions such as car driving.Across the three DORAs, driving ability was indeed impaired in some subjects in dedicated driving studies in which driving performance was evaluated by means of keeping a steady lane position during a long monotonous highway drive of 1 h, which represents a challenge of sustained vigilance and attention (Muehlan et al., 2022;Vermeeren et al., 2015Vermeeren et al., , 2019)).Assessment of driving performance is an integral part of FDA's safety evaluation of new drugs that have a pronounced effect on the CNS, and these driving studies can either be performed during a real drive on the road or in a simulator (FDA, 2017).Drugs that affect driving performance may also reduce a person's ability to recognise and judge his or her own impairment.Therefore, it is important to assess objective driving performance (e.g., keeping in lane) in combination with self-rated judgement of the quality of the drive (i.e., subjective performance) to better quantify the degree of impairment and to identify and mitigate potential risks caused by unawareness of impairment or overestimation of driving skills (Verster & Roth, 2011, 2012).Interestingly, some studies show that after completion of the driving test when treated with a DORA, participants were able to perceive their driving difficulties (Muehlan et al., 2022;Vermeeren et al., 2016), in contrast to reports with other CNS-active products (e.g., GABAergic drugs) that may distort the perception of driving difficulties (Verster & Roth, 2012).
The maintained perception of impaired driving observed with DORAs may provide a safeguard.As impairment of driving performance was present in some subjects, the prescribing information of DORAs warns about the increased risk of daytime impairment if taken with less than a full night of sleep remaining, and additional cautionary statements are indicated towards driving and other activities requiring complete mental alertness (FDA, 2014(FDA, , 2019(FDA, , 2022)).
The tolerability profile of the respective DORAs is heterogeneously reported.The most commonly observed AE, reported at least twice the incidence of placebo, was dose-dependent somnolence with suvorexant and lemborexant (FDA, 2014(FDA, , 2019)).At the high dose of 20 mg, 5%-8% of patients treated with suvorexant versus 3% on placebo reported somnolence (Herring et al., 2016).With lemborexant at the approved top dose of 10 mg, 7% of patients versus 2% on placebo in Study 1 (Rosenberg et al., 2019), and 13% on lemborexant versus 2% on placebo in the long-term Study 2 (Kärppä et al., 2020) reported somnolence.The most commonly reported AE with daridorexant was headache, with 6% of the subjects at the high dose versus 4% on placebo, while somnolence was reported in 2% of the subjects at the high dose of 50 mg with no dose dependency and no apparent difference to the placebo group (Mignot et al., 2022).Notably, as per USA prescribing information, both suvorexant (10 mg) and lemborexant (5 mg) have a recommended starting dose that may be increased based on clinical response and tolerability, whereas daridorexant can be initiated at either approved dose, 25 or 50 mg.In Europe, daridorexant is the first-in-class DORA with a recommended dose of 50 mg, based on the superior efficacy/improved daytime functioning compared to 25 mg, with no apparent trade-off in safety (EMA, 2022).
The safety and tolerability profile of older hypnotic classes has hampered the use of pharmacological interventions in situations in which treatment modalities were limited or inadequate (e.g., vulnerable populations) and has left these patients in a dilemma of increased risks using, e.g., a benzodiazepine and their potential benefits.In 2019, the American Geriatrics Society Beers Criteria recommended that Z-drugs should be avoided in older adults due to AEs similar to those of benzodiazepines (AGS, 2019).This is in marked contrast to the DORAs that have shown a positive benefit-risk profile also in the elderly population (Fietze et al., 2022;Herring et al., 2017;Rosenberg et al., 2019).The overall safety and tolerability profile of DORAs in elderly patients was generally consistent with younger patients, and no dose adjustments based on age are required (Fietze et al., 2022;Moline et al., 2020;Murphy, Moline et al., 2017;Zammit et al., 2020).
Also, the use of benzodiazepines or Z-drugs is of particular concern in patients with compromised respiratory function, as the drugs are associated with respiratory depression (Rosenberg et al., 2021).The safety of DORAs has been investigated in patients with chronic obstructive pulmonary disease (COPD) and obstructive sleep apnoea (OSA) of different severity grades, in which the prevalence of insomnia symptoms is high (Boof et al., 2020;Boof et al., 2021;Boof et al., 2022;Cheng et al., 2020;Sun et al., 2016).In contrast to benzodiazepines and their association with respiratory depression, results showed no clinically meaningful effect on night-time respiratory function (assessed by the apnoea-hypopnoea index and peripheral oxygen saturation), confirmed by the absence of any respiratory-related AEs when DORAs were administered in subjects with mild-moderate OSA or moderate COPD.Acknowledging the limitations of these safety studies (e.g., small sample size and a relatively short treatment duration of 4-8 days), the DORAs did not impair night-time respiratory function.
The DORAs exert their activity on the CNS.Therefore, in accordance with the United States Controlled Substances Act, they were all evaluated in a large preclinical and clinical programme to characterise their abuse and dependence potential.This includes evaluation of animal behavioural pharmacology and a dedicated human abuse potential (HAP) study in healthy participants with a history of using CNS depressants for recreational purpose (Landry et al., 2022;Schoedel et al., 2016;Ufer et al., 2022).Summarised, there was a lack of positive reinforcing effects of the DORAs in self-administration studies in rats, which is in sharp contrast to drugs of the benzodiazepine or Z-drug classes, which have been shown to maintain selfadministration in animals (Asakura et al., 2021;Roch et al., 2021;Steiner et al., 2013;Ufer et al., 2020).The HAP studies are mainly based on evaluation of self-rated visual analogue scales, on which DORAs showed greater drug-liking than placebo, specifically at high supratherapeutic doses.The overall AE profile in these studies was in line with the known pharmacology of these drugs following morning  et al., 2016;Dauvilliers et al., 2020;Pépin et al., 2021;Schweitzer et al., 2019), OSA may be a new therapeutic target for the DORA class.
Patients with substance use disorders (SUDs) often have concomitant insomnia symptoms (NIDA, 2018).These patients should be kept abstinent from the abused substance (e.g., alcohol), and GABA A agonists should be avoided if concomitant insomnia symptoms are to be treated.However, if left untreated, insomnia has been associated with an increased risk of relapse in these subjects and DORAs might be a therapeutic opportunity, based on first results in small cohorts that need to be confirmed in larger trials (Campbell et al., 2020;Geoffroy et al., 2020).The hypothesis to treat patients with insomnia and SUDs with a DORA is further supported by their inherently low abuse potential compared to benzodiazepines and Z-drugs (James et al., 2020;Matzeu & Martin-Fardon, 2021;Steiner et al., 2013).
SUDs of interest with a clear medical need include alcohol use disorder (Campbell et al., 2018(Campbell et al., , 2020;;Walker & Lawrence, 2017) and opioid use disorder (Greenwald et al., 2021;James et al., 2020;Matzeu & Martin-Fardon, 2021).With respect to the latter, a recent small study funded by the United States National Institute on Drug Abuse investigated the clinical utility of suvorexant in subjects undergoing stepwise buprenorphine/naloxone tapering as treatment of opioid withdrawal.
Results are remarkable in that suvorexant was well tolerated and improved TST, self-reported opioid-use withdrawal symptoms, and craving (Huhn et al., 2022).Multiple clinical trials are ongoing to further investigate the benefit of add-on suvorexant to improve sleep and opioid abstinence, stress, and compliance/relapse in patients with SUDs (NCT04262193, NCT05546515, NCT05145764, NCT04287062).
The overall preservation of sleep architecture in animals is a unique feature of the DORA class compared to GABA A modulators (Roch et al., 2021;Tannenbaum et al., 2016).In man, based on pooled analyses of sleep studies with suvorexant and lemborexant, sleep efficiency and TST improved, with a proportional increase in the time spent in all sleep stages (Mignot et al., 2022;Moline, Zammit, Cheng et al., 2021;Snyder et al., 2016), with no increase in abnormal REMsleep behaviour.Future studies are needed to further elucidate the clinical relevance of maintaining natural sleep architecture in patients treated with DORAs.
Regarding the use of orexin agonists for the treatment of narcolepsy, promising first results in animals and humans (Evans et al., 2022(Evans et al., , 2023) ) will need to be confirmed in larger trials to shed light on the question of whether these drugs can improve daytime alertness, suppress cataplexy, and overall stabilise the wake and sleep rhythm of patients with narcolepsy.
An ongoing study is investigating whether daridorexant improves sleep in patients with insomnia and comorbid nocturia based on changes in sTST assessed with sleep diaries (NCT05597020).
Extensive research is also ongoing outside the field of insomnia, which is not the topic of this review.Considering the increasing body of informative data, the range of potential applications of treatments mediating orexin signalling is only limited by the imagination of the researchers.

F
I G U R E 1 Schematic representation of the stabilising effect of orexin on wakefulness and model of insomnia/hyperarousal.(a) During wakefulness, the monoaminergic nuclei (yellow) inhibit the VLPO (blue), thereby relieving the inhibition of the monoaminergic cells, and that of the orexin neurones (green).(b) During sleep, VLPO neurones inhibit the monoaminergic cell groups, thereby relieving their own inhibition.In addition, orexin neurones are inhibited, further preventing monoaminergic activation that might interrupt sleep.The mutual inhibition between the VLPO and the monoaminergic nuclei produces sharp transitions of sleep-wake state.Orexins stabilise wakefulness.(c) Insomnia is described as a state of hyperarousal.Inappropriately high orexin levels are thought to contribute to the hyperarousal state of insomnia.DORAs inhibit orexin receptors, alleviating the potential hyperarousal effect of orexins.eVLPO, extended ventrolateral preoptic nucleus; DORA, dual orexin receptor antagonist; LC, locus coeruleus; TMN, tuberomammillary nucleus; VLPO, ventrolateral preoptic nucleus (Graphic adapted from Saper et al., 2005, Figure 4).[Color figure can be viewed at wileyonlinelibrary.com] . Prior to the introduction of the DORA class, pharmacological interventions with a clinically meaningful improvement demonstrated in large randomised controlled trials of multiple night-time symptoms experienced by patients (such as sleep onset and sleep maintenance covering the whole night), were F I G U R E 2 Maximum plasma concentration (C max )-normalised concentration-time profiles of suvorexant, lemborexant, and daridorexant (%).The data on the basis of which these profiles were constructed are taken from FDA (2013) for suvorexant, Landry et al. (2020) for lemborexant, Muehlan et al. (2018) for daridorexant.[Color figure can be viewed at wileyonlinelibrary.com] desperately lacking.The DORAs have all shown clinical utility in improving sleep onset and sleep maintenance endpoints in robust clinical trials

a
Treatment effects statistically significantly superior ( p < 0.05) to placebo after multiplicity adjustment.Although multiple time points have been used in the testing strategy of the key endpoint, the longest treatment duration is represented.may contribute to improvement in daytime functioning(Levenson et al., 2015;Smith et al., 2015), the relationship between night-time and daytime symptoms has not been elucidated, nor genuinely addressed during drug development.Therefore, in contrast to the known effect on sleep, the impact of insomnia treatments on daytime functioning until recently remained largely unexplored.Mignot et al. (2022) reported encouraging data on improved daytime functioning and sleep outcomes observed with a DORA.Using the Insomnia Daytime Symptoms and Impacts Questionnaire (IDSIQ), a self-reporting instrument validated in patients with DSM-5 defined insomnia(Hudgens et al., 2021), daridorexant 50 mg showed a statistically and clinically meaningful improvement of daytime symptoms of insomnia.

T A B L E 3
Efficacy of lemborexant 5 and 10 mg in Study 2 at Month 1: effects on objective sleep variables in patients with insomnia disorder.

T A B L E 5
Efficacy of daridorexant 25 and 50 mg at Month 3: effects on objective sleep variables in patients with insomnia disorder.À24.3 (À29.0 to À19.5) À10.3 (À17.0 to À3.5) Placebo 108.1 (48.7)À14.0 (À18.8 to À9.2) -Note: sample size for Study 1 was 1310 participants each for daridorexant 25 mg, 50 mg, and placebo.Sample size for Study 2 was 309 participants for daridorexant 25 mg and 308 participants for placebo.Abbreviations: CI, confidence interval; LPS, latency to persistent sleep; LSM, least squares means; SD, standard deviation; WASO, wake after sleep onset.a Treatment effects statistically significantly superior ( p < 0.05) to placebo after multiplicity adjustment.Although multiple time points have been used in the testing strategy of the key endpoint, the longest treatment duration is represented.TA B L E 6 Efficacy of daridorexant 25 and 50 mg at Month 3: effects on subjective sleep variables in patients with insomnia disorder.
Beneficial effects of DORAs on sleep onset and maintenance have been demonstrated.This new class of prescription therapies improves sleep variables with fewer side effects than the older pharmacological approaches.Based on recent evidence, improvements in sleep outcomes may be achieved with a lower risk of next-morning sleepiness and even improvements in daytime functioning.The combination of both improved night-time sleep and daytime alertness may lead to future research targeting not only depression, AD, SUDs, psychiatric, or paediatric neurodevelopmental disorders, but also conditions secondary to insomnia, such as impaired cognitive function.
Note: dosing: 15 mg in elderly aged ≥ 65 years, 20 mg in individuals aged ≥18-64 years.For the two suvorexant studies (Tables1 and 2), 291 non-elderly adults were treated with 20 mg and 449 with placebo, while 202 elderly patients were treated with 15 mg and 318 with placebo.
Abbreviations: CI, confidence interval; LPS, latency to persistent sleep; LSM, least squares means; SD, standard deviation; WASO, wake after sleep onset.a Treatment effects statistically significantly superior ( p < 0.05) to placebo after multiplicity adjustment.Although multiple time-points have been used in the testing strategy of the key endpoint, the longest treatment duration is represented.T A B L E 2 Efficacy of suvorexant 15/20 mg at Month 3: effects on subjective sleep variables in patients with insomnia disorder.Note: dosing: 15 mg in elderly aged ≥ 65 years, 20 mg in individuals aged ≥18-64 years.For the two suvorexant studies (Tables Note: sample size was 260 participants for lemborexant 5 mg, 260 participants for lemborexant 10 mg, and 200 participants for placebo.Abbreviations: CI, confidence interval; LPS, latency to persistent sleep; LSGM, least square geometric mean; LSM, least squares means; SD, standard deviation; SEF, sleep efficiency; WASO, wake after sleep onset.Treatment effects statistically significantly superior ( p < 0.05) to placebo after multiplicity adjustment.Efficacy of lemborexant 5 and 10 mg in Study 1 at Month 6: effects on subjective sleep variables in patients with insomnia disorder.
a T A B L E 4Note: sample size was 316 participants for lemborexant 5 mg, 315 participants for lemborexant 10 mg, and 318 participants for placebo.Abbreviations: LSM, least squares means; Q, quartile; SD, standard deviation; SE, standard error; sSEF, subjective sleep efficiency; sSOL, subjective sleep onset latency; sWASO, subjective wake after sleep onset.a Treatment effects statistically significantly superior ( p < 0.05) to placebo after multiplicity adjustment.
Treatment effects statistically significantly superior ( p < 0.05) to placebo after multiplicity adjustment.Although multiple time points have been used in the testing strategy of the key endpoint, the longest treatment duration is represented. a